A 3D printer printing failure detection method and system

Abstract

The present invention relates to the technical field of 3D printers, and particularly to a method and system for detecting 3D printer printing failures. The method includes the following steps: by obtaining the 3D printer operation log, extracting the abnormal state data of fused deposition modeling, and then inversely deducing the abnormal wire drawing mechanism and estimating the probability of warping between printing layers; then, performing a cracking risk calculation on the warping probability data to obtain the cracking risk data between printing layers; based on a convolutional neural network, constructing a cracking risk identification model between printing layers, and sending the model to the 3D printer control center for performing printing failure detection. The present invention makes the 3D printer printing failure detection technology more perfect through the optimization of the 3D printing failure detection technology.

Description

Technical Field

[0001] The present invention relates to the technical field of 3D printers, and particularly to a method and system for detecting 3D printer printing failures. Background Art

[0002] With the rapid development of 3D printing technology, 3D printers have been widely used in fields such as manufacturing, medical, aerospace, and automotive. As an additive manufacturing method of layer-by-layer stacking, 3D printing has high degrees of freedom and design flexibility and can manufacture objects with complex shapes. However, in practical applications, various problems often occur during the 3D printing process, resulting in printing failures, such as warping, cracking, peeling, and defects between printing layers. These problems not only affect the printing effect but also lead to waste of raw materials, extended printing time, and even damage to the printing equipment. Therefore, how to detect and predict abnormal states during the printing process in real time and timely discover potential printing problems has become the key to improving 3D printing quality, reducing production costs, and increasing efficiency. In the FDM printing process, abnormal filament drawing is one of the key factors affecting printing quality. Abnormal filament drawing usually occurs after the nozzle extrudes the material. Due to the uneven cooling of the molten thermoplastic filamentous material in the air, the filamentous material fails to be deposited as expected but instead exhibits problems such as filamentous stretching, stacking, or adhesion. This abnormal filament drawing not only affects the printing surface quality but also causes uneven adhesion between printing layers, further leading to warping problems. Warping between layers is due to differences in shrinkage rates during the material cooling process, especially when the cooling rates of different layers are inconsistent, the interlayer adhesion force decreases, resulting in warping. Warping not only affects the dimensional accuracy of the printed part. When the warping degree is large or the interlayer bonding force of the material is insufficient, a stable bond cannot be formed between different layers of the printed part, ultimately leading to interlayer separation or obvious cracks, ultimately triggering the risk of cracking. However, traditional methods for detecting 3D printer printing failures have problems such as low risk recognition accuracy for abnormal states (such as filament drawing, warping, and cracking), weak ability to trace the causes of abnormal states, resulting in large errors in printing failure detection. Summary of the Invention

[0003] Based on this, it is necessary to provide a method and system for detecting 3D printer printing failures to solve at least one of the above technical problems.

[0004] To achieve the above object, a method for detecting 3D printer printing failures includes the following steps:

[0005] Step S1: Obtain the 3D printer operation log; extract the fused deposition modeling abnormal state from the 3D printer operation log to obtain the fused deposition modeling abnormal state data;

[0006] Among them, step S2 includes:

[0007] Step S21: Obtain the status data of the thermoplastic filament material;

[0008] Step S22: Based on the abnormal status data of fused deposition modeling, identify the abnormal printing morphology deformation to obtain the abnormal printing morphology deformation data;

[0009] Step S23: According to the status data of the thermoplastic filament material, inversely deduce the abnormal wire drawing mechanism for the abnormal printing morphology deformation data to obtain the abnormal wire drawing mechanism;

[0010] Step S24: Based on the abnormal printing morphology deformation data and the abnormal wire drawing mechanism, estimate the probability of warping between printing layers to obtain the probability data of warping between printing layers;

[0011] Step S25: Perform a cracking risk calculation on the probability data of warping between printing layers to obtain the cracking risk data between printing layers;

[0012] Step S3: Based on a convolutional neural network, construct a cracking risk identification model for the cracking risk data between printing layers to obtain a cracking risk identification model for the printing layers;

[0013] Step S4: Send the cracking risk identification model for the printing layers to the 3D printer control center to perform printing failure detection.

[0014] Preferably, step S1 includes the following steps:

[0015] Step S11: Obtain the 3D printer operation log;

[0016] Step S12: Clean the data of the 3D printer operation log to obtain the cleaned 3D printer operation log;

[0017] Step S13: Extract the abnormal status of fused deposition modeling from the cleaned 3D printer operation log to obtain the abnormal status data of fused deposition modeling.

[0018] Preferably, step S23 includes the following steps:

[0019] Step S231: Analyze the thermal viscosity drop gradient of the status data of the thermoplastic filament material to obtain the thermal viscosity drop gradient data;

[0020] Step S232: Based on the thermal viscosity drop gradient data, simulate the change in the thermoplastic fracture elongation rate of the status data of the thermoplastic filament material to obtain the change data of the thermoplastic fracture elongation rate;

[0021] Step S233: Analyze the abnormal deformation contour parameters of the abnormal printing morphology deformation data to obtain the abnormal deformation contour parameters;

[0022] Step S234: Perform local abnormal deformation thermoplastic elongation rate distribution inversion on the abnormal deformation profile parameters according to the thermoplastic fracture elongation rate change data to obtain the abnormal deformation thermoplastic distribution elongation rate;

[0023] Step S235: Perform abnormal printing retraction distance / speed calculation based on the abnormal deformation thermoplastic distribution elongation rate and the abnormal deformation profile parameters to obtain the abnormal printing retraction distance / speed;

[0024] Step S236: Reverse-infer the abnormal wire drawing mechanism for the abnormal deformation profile parameters according to the abnormal printing retraction distance / speed and the abnormal deformation thermoplastic distribution elongation rate to obtain the abnormal wire drawing mechanism.

[0025] Preferably, step S24 includes the following steps:

[0026] Step S241: Identify the internal cavity distribution of the printing abnormal shape deformation data to obtain the abnormal shape internal cavity distribution data;

[0027] Step S242: Reverse-infer the change in material hot melt tension based on the abnormal wire drawing mechanism to obtain the material hot melt tension change data;

[0028] Step S243: Perform material hot melt cooling shrinkage rate increment gradient calculation according to the abnormal shape internal cavity distribution data and the material hot melt tension change data to obtain the hot melt cooling shrinkage rate increment gradient;

[0029] Step S244: Perform multiple linear regression analysis on the hot melt cooling shrinkage rate increment gradient to obtain the shrinkage rate increment linear regression gradient;

[0030] Step S245: Estimate the probability of warping between printing layers based on the shrinkage rate increment linear regression gradient to obtain the probability data of warping between printing layers.

[0031] Preferably, step S25 includes the following steps:

[0032] Step S251: Evaluate the non-uniform cooling rate of the printing layer warping probability data to obtain the non-uniform cooling rate of layer warping;

[0033] Step S252: Perform numerical fluctuation simulation of internal stress accumulation on the printing layer warping probability data according to the non-uniform cooling rate of layer warping to obtain the numerical fluctuation data of interlayer internal stress;

[0034] Step S253: Analyze the risk interval of interlayer adhesion instability based on the numerical fluctuation data of interlayer internal stress and the non-uniform cooling rate of layer warping to obtain the risk interval of interlayer adhesion instability;

[0035] Step S254: Perform cracking risk calculation according to the risk interval of interlayer adhesion instability to obtain the cracking risk data of printing layers.

[0036] Preferably, step S3 includes the following steps:

[0037] Step S31: Perform convolution calculation on the risk level of interlayer cracking during printing to obtain convolution data of the risk of interlayer cracking during printing;

[0038] Step S32: Perform logical learning on the convolution data of the risk of interlayer cracking during printing to obtain logical data of the risk of interlayer cracking during printing;

[0039] Step S33: Based on a convolutional neural network, construct a risk identification model for interlayer cracking during printing from the logical data of the risk of interlayer cracking during printing, to obtain a risk identification model for interlayer cracking during printing.

[0040] Preferably, step S33 includes the following steps:

[0041] Step S331: Extract multi-scale features from the logical data of the risk of interlayer cracking during printing to obtain multi-scale risk data of interlayer cracking;

[0042] Step S332: Divide the multi-scale risk data of interlayer cracking into a training set and a test set, respectively obtaining a multi-scale risk training set of interlayer cracking and a multi-scale risk test set of interlayer cracking;

[0043] Step S333: Based on a convolutional neural network, construct an initial risk identification model for interlayer cracking during printing from the multi-scale risk training set of interlayer cracking, to obtain an initial risk identification model for interlayer cracking during printing;

[0044] Step S334: Test the initial risk identification model for interlayer cracking during printing according to the multi-scale risk test set of interlayer cracking, to obtain a risk identification model for interlayer cracking during printing.

[0045] Preferably, the present invention also provides a 3D printer printing failure detection system for executing the 3D printer printing failure detection method as described above. The 3D printer printing failure detection system includes:

[0046] An abnormal state extraction module, configured to obtain the operation log of the 3D printer; extract the abnormal state of fused deposition modeling from the operation log of the 3D printer to obtain abnormal state data of fused deposition modeling;

[0047] A cracking risk calculation module, configured to perform reverse deduction of the abnormal wire drawing mechanism based on the abnormal state data of fused deposition modeling to obtain the abnormal wire drawing mechanism; estimate the probability of interlayer warping during printing based on the abnormal wire drawing mechanism to obtain probability data of interlayer warping during printing; perform cracking risk calculation on the probability data of interlayer warping during printing to obtain cracking risk data of interlayer cracking during printing;

[0048] The cracking risk identification model construction module is used to construct a printing layer cracking risk identification model based on a convolutional neural network for the printing layer cracking risk data, and obtain the printing layer cracking risk identification model;

[0049] The detection execution module is used to send the printing layer cracking risk identification model to the 3D printer control center to perform printing failure detection.

[0050] The beneficial effect of the present invention is that by obtaining the operation log of the 3D printer, the system can record various key parameters in the printing process in real time, such as temperature, printing speed, nozzle status, etc. These data provide valuable information resources for subsequent analysis. By extracting the abnormal state of fused deposition modeling from the operation log, abnormal conditions in the printing process, such as abnormal temperature fluctuations and nozzle blockage, can be accurately identified. This process not only helps to capture potential printing problems in a timely manner, but also provides an accurate data basis for subsequent mechanism analysis and risk prediction, avoiding omissions and delays in traditional manual monitoring. Based on the extracted abnormal state data of fused deposition modeling, the system reversely infers the abnormal wire drawing mechanism and analyzes the warping phenomenon between printing layers caused by uneven wire drawing, overheating of materials, etc. during the printing process. These abnormal wire drawing mechanisms reflect the root causes of physical deformation and defects in the printing process. Based on these mechanisms, the system further estimates the probability of warping between printing layers, and calculates the cracking risk through the cracking risk calculation model. This step can effectively predict structural problems in the printing process, provide a basis for subsequent measures, and thus improve the printing success rate. Through deep learning analysis of the interlayer cracking risk data using a convolutional neural network (CNN), the system can automatically identify the risk of interlayer cracking during the printing process. In this process, the CNN model extracts potential patterns and rules by learning a large amount of abnormal data, and then establishes an accurate cracking risk identification model. This model has a strong generalization ability and can accurately identify the risk of cracking under different printing conditions and environments. This step not only reduces the need for manual intervention, but also improves the accuracy and efficiency of risk identification, and timely detects problems and issues early warning. The constructed interlayer cracking risk identification model is sent to the 3D printer control center in real time. This operation enables the printer to obtain risk identification results in real time during the printing process and take corresponding preventive measures based on the risks predicted by the model. For example, the control center can automatically correct the problem that causes cracking by adjusting printing parameters (such as temperature, speed, path, etc.). This closed-loop control method greatly improves the adaptability and intelligence level of the printing process, making printing failure detection and risk management more efficient and accurate, greatly reducing the probability of printing failure, and ensuring the stability of the printing process and the quality of the final product. In the existing technology, the independent identification of wire drawing, edge warping and cracking has two key defects: lack of contextual association, inability to understand the evolution relationship between defects, and high false alarm rate: single feature recognition is easily affected by environmental interference and random factors. This case understands the complete evolution process from wire drawing to edge warping to cracking, effectively reducing the false alarm rate and improving recognition accuracy.Therefore, the present invention is an optimization of the traditional 3D printer printing failure detection method, which solves the problems of the traditional 3D printer printing failure detection method, such as low risk identification accuracy for abnormal states (such as wire drawing, warping, and cracking), weak ability to trace the causes of abnormal states, resulting in large errors in printing failure detection. It improves the risk identification accuracy for abnormal states (such as wire drawing, warping, and cracking) and the ability to trace the causes of abnormal states, and reduces the errors in printing failure detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is a schematic diagram of the step flow of a 3D printer printing failure detection method;

[0052] Figure 2 For Figure 1 it is a schematic diagram of the detailed implementation step flow of step S2 in

[0053] Figure 3 For Figure 1 it is a schematic diagram of the detailed implementation step flow of step S3 in DETAILED DESCRIPTION OF THE INVENTION

[0054] Please refer to Figures 1 to 3 , a 3D printer printing failure detection method, the method includes the following steps:

[0055] Step S1: Obtain the 3D printer operation log; extract the fused deposition modeling abnormal state from the 3D printer operation log to obtain the fused deposition modeling abnormal state data;

[0056] Step S2: Inverse-deduce the abnormal wire drawing mechanism based on the fused deposition modeling abnormal state data to obtain the abnormal wire drawing mechanism; estimate the warping probability between printing layers based on the abnormal wire drawing mechanism to obtain the warping probability data between printing layers; perform a cracking risk calculation on the warping probability data between printing layers to obtain the cracking risk data between printing layers;

[0057] Step S3: Build a printing layer cracking risk identification model based on the convolutional neural network for the printing layer cracking risk data to obtain the printing layer cracking risk identification model;

[0058] Step S4: Send the printing layer cracking risk identification model to the 3D printer control center to perform printing failure detection.

[0059] In the embodiment of the present invention, referring to Figure 1 as described, it is a schematic diagram of the step flow of a 3D printer printing failure detection method of the present invention. In this example, the 3D printer printing failure detection method includes the following steps:

[0060] Step S1: Obtain the running log of the 3D printer; extract the abnormal state of fused deposition modeling from the running log of the 3D printer to obtain the abnormal state data of fused deposition modeling;

[0061] In the embodiments of the present invention, it is first necessary to obtain the running log data of the 3D printer. The running log of the printer is generally recorded by multiple sensors and control systems, covering key data such as the temperature of the printer, the state of the extruder, the nozzle movement trajectory, the material deposition situation, and the cooling rate. The obtained log data is usually stored in a time series manner and contains abnormal alarm information during the operation of the device. This log data can be exported from the printer control system through a serial communication interface or a network communication protocol (such as MQTT, HTTP API) and stored in a local database or a log analysis system for processing. When preprocessing the running log data of the 3D printer, data cleaning and format conversion are required. The main purpose of data cleaning is to remove redundant data, fill in missing values, unify the data format, and screen out the data fields related to the abnormal state of fused deposition modeling. The cleaned data needs to be subjected to abnormal state extraction. The specific method is to calculate the change trend of the data based on time series anomaly detection algorithms, such as the sliding window statistical method or the method based on dynamic time warping (DTW). For example, during the fused deposition process, the temperature of the nozzle should generally be maintained within the range of the set value ±2°C. If the temperature deviation at a certain time point exceeds this range and lasts for a period of time, it is determined that a temperature anomaly has occurred during this period. In addition, abnormal detection is also required for parameters such as the extrusion speed and interlayer bonding pressure of the printer. For example, the Mahalanobis distance or the local outlier factor (LOF) method is used to identify outliers in the multi-dimensional sensor data, and finally the abnormal state data of fused deposition modeling is obtained.

[0062] Step S2: Based on the abnormal state data of fused deposition modeling, reverse-deduce the abnormal wire drawing mechanism to obtain the abnormal wire drawing mechanism; estimate the warping probability between printing layers based on the abnormal wire drawing mechanism to obtain the warping probability data between printing layers; perform a cracking risk calculation on the warping probability data between printing layers to obtain the cracking risk data between printing layers;

[0063] In the embodiments of the present invention, based on the abnormal state data of fused deposition modeling, it is necessary to further deduce the abnormal wire drawing mechanism. Abnormal wire drawing mainly occurs at the extrusion end and between printing layers. Due to the different rheological properties of the material at different temperatures and extrusion speeds, abnormal wire drawing can be attributed to the viscosity change of the hot-melt filamentous material and the retraction behavior of the extruder under specific working conditions. First, by obtaining the state data of the hot-melt filamentous material, including parameters such as its melt flow index (MFI), viscosity curve, and tensile properties, and combining the nozzle temperature change curve and extrusion rate fluctuation data measured in the experiment, analyze the viscosity decrease gradient of the material at different temperatures. During the experimental measurement, use a high-precision thermocouple to monitor the nozzle outlet temperature and record the material fluidity curve at different temperatures. Secondly, based on the thermal viscosity decrease gradient of the material, use the finite element analysis (FEA) method to simulate the change of the thermoplastic fracture elongation rate of the material and calculate the melt index increment of the material at a specific temperature. For example, if the MFI of a certain PLA material is 6 g / 10 min at 190°C and increases to 10 g / 10 min at 210°C, it can be inferred that in the range of 190°C to 210°C, the fluidity of the material increases sharply, resulting in an increased probability of material wire drawing during printing. By analyzing the retraction distance and speed of the extruder, use partial differential equations (PDEs) to describe the flow behavior of the material, and correct the model in combination with experimental data to finally obtain the abnormal wire drawing mechanism. The abnormal wire drawing mechanism directly affects the probability of warping between printing layers. Therefore, it is necessary to estimate the probability of warping between layers based on the abnormal shape deformation data and the change data of the hot-melt tension of the material. First, analyze the internal cavity distribution of the abnormal shape deformation data, and use X-ray CT scanning technology or voxel reconstruction method based on optical measurement to extract the distribution characteristics of the cavities between printing layers. For example, in the printing experiment of a certain ABS material, if the interlayer bubble rate exceeds 1.5%, the probability of warping increases significantly. Based on the abnormal wire drawing mechanism, calculate the change of the hot-melt tension formed during the wire drawing process of the material, and use the finite element simulation or experimental measurement method to analyze the increment gradient of the cooling shrinkage rate of the material at different tensile speeds. For example, use a high-precision displacement sensor to measure the shrinkage rate of the material after cooling and record the shrinkage rate change under different temperature conditions to obtain the shrinkage rate increment curve. Based on these data, fit the shrinkage rate increment gradient through multiple linear regression analysis and calculate the probability of warping between printing layers using the Bayesian probability model to obtain the probability data of warping between layers. When calculating the cracking risk of the probability data of warping between printing layers, it is necessary to calculate in combination with the non-uniform cooling rate of the material and the interlayer stress accumulation. First, use an infrared thermal imager or thermocouple array to monitor the cooling process of the printed part and obtain the interlayer cooling rate distribution data. For example, when printing ABS material, if the interlayer cooling rate is in the range of 3°C / s to 5°C / s, the interlayer stress is low, while if the cooling rate exceeds 10°C / s, the interlayer stress accumulation increases significantly.Based on the cooling rate data, the numerical fluctuation of the interlayer internal stress is calculated by using the numerical simulation method, and the risk interval of the interlayer adhesion instability is identified through the distribution of monotonically increasing mutation points. The Newton iteration method is used to numerically solve the mutation points of the internal stress, and combined with the thermal stress analysis model, the cracking risk data of the printing interlayer is obtained. It should be noted that the main reasons for abnormal wire drawing leading to warping and cracking are the hot melt rheological properties and cooling shrinkage behavior of the material. In the fused deposition modeling process, first, a hot melt filamentous material is extruded. If the extrusion temperature fluctuates or the extrusion rate is unstable, the hot melt viscosity of the material decreases, the printing retraction is abnormal, or the cooling rate is uneven, and then slender wire drawing is formed on the printing path. These wire drawings not only affect the uniform deposition of the interlayer material but also form non-uniform thermal shrinkage in local areas. Since the material will undergo volume shrinkage during the cooling process, the non-uniform deposition caused by abnormal wire drawing will cause changes in the local cooling rate, and then lead to non-uniform distribution of the interlayer internal stress. When the internal stress accumulation exceeds the interlayer adhesion strength of the material, the warping phenomenon will occur. Especially in a high cooling rate environment, the thermal shrinkage stress increases, making the warping trend more obvious. At the same time, excessive local interlayer stress will form microcracks and further expand during the subsequent printing process, ultimately causing interlayer cracking.

[0064] Step S3: Based on the convolutional neural network, construct a printing interlayer cracking risk identification model for the printing interlayer cracking risk data to obtain the printing interlayer cracking risk identification model;

[0065] In the embodiment of the present invention, based on the printing interlayer cracking risk data, a printing interlayer cracking risk identification model is constructed by using a convolutional neural network (CNN). First, multi-scale feature extraction is performed on the printing interlayer cracking risk data, including local stress distribution, material shrinkage rate gradient, printing path influence factor, etc. The sliding window technique is used to preprocess the data, and a suitable convolutional kernel size (such as 3×3 or 5×5) is selected to extract risk features of different scales. Then, the data is divided into a training set and a test set, and the cross-validation method is used to evaluate the model performance. During the training process, the Adam optimizer is used to adjust the learning rate, and the model parameters are optimized according to the loss function (such as mean square error or cross-entropy loss). Finally, the generalization ability of the model is verified on the test set, and the hyperparameters are adjusted to obtain the final printing interlayer cracking risk identification model.

[0066] Step S4: Send the printing interlayer cracking risk identification model to the 3D printer control center to perform printing failure detection.

[0067] In the embodiment of the present invention, the trained printing layer cracking risk identification model is deployed to the 3D printer control center and integrated with the real-time monitoring system of the printer. By collecting the operation data of the printer in real time and inputting it into the cracking risk identification model, the failure probability under the current printing state is calculated. If it is detected that the cracking risk exceeds the threshold (such as 90%), the printer control system can trigger an alarm and take measures such as adjusting the printing parameters, reducing the printing speed, or increasing the extrusion temperature to reduce the printing failure risk.

[0068] Step S1 includes the following steps:

[0069] Step S11: Obtain the 3D printer operation log;

[0070] Step S12: Clean the data of the 3D printer operation log to obtain the 3D printer operation cleaning log;

[0071] Step S13: Extract the fused deposition modeling abnormal state from the 3D printer operation cleaning log to obtain the fused deposition modeling abnormal state data.

[0072] In the embodiments of the present invention, the process of obtaining the 3D printer operation log includes multiple key links such as data acquisition, storage, and time synchronization. The printer's control system records the device operation status in real time through built-in sensors and control modules, and stores the relevant information in the form of a structured log file. The operation log usually includes parameters such as nozzle temperature, extruder status, printing speed, platform temperature, Z-axis height, material flow rate, and cooling fan speed, and these parameters are all timestamped. The data acquisition method mainly extracts data from the printer's firmware through serial communication interfaces (such as UART, RS-232) or network communication protocols (such as MQTT, HTTP API), and stores it in a local database or log file. The data storage format generally uses CSV, JSON, or binary format for subsequent data processing. When obtaining the log, it is necessary to ensure log time synchronization for subsequent anomaly analysis. For example, use NTP (Network Time Protocol) or GPS time synchronization technology to ensure that all log data is recorded according to a unified time reference, avoiding anomaly detection errors caused by time offsets. The obtained operation log usually contains a large amount of redundant data, missing values, and inconsistent formats, so it is necessary to clean the data to ensure the accuracy of subsequent analysis. First, standardize the format of the log data, unify the field names, data types, and units, such as unifying the temperature unit to degrees Celsius (°C) and converting the time format to the ISO 8601 standard format. Secondly, use interpolation methods to fill in the missing data. For the problem of missing parameters caused by communication loss during the printing process, use linear interpolation, spline interpolation, or locally weighted regression methods to reconstruct the data. For example, if the nozzle temperature data at a certain time point is missing, linear interpolation calculation can be performed based on the temperature values at the previous and subsequent time points. Then, eliminate the abnormal data. Use box plot analysis (IQR) or the 3σ principle to eliminate outliers. For example, when the number of pulses of the extruder stepper motor fluctuates more than ±5σ of the average value in a short period of time, this data point can be regarded as abnormal data and deleted. Finally, perform data denoising processing. Use moving average filtering or Kalman filtering to remove sensor noise. For example, use data points with a window size of 5 to calculate the moving average to smooth the temperature change curve. The cleaned log data is stored in a new data table or file for subsequent extraction of abnormal states in fused deposition modeling. The process of extracting abnormal states in fused deposition modeling from the operation log involves steps such as multi-dimensional feature analysis, time series anomaly detection, and pattern matching. First, extract the key process parameters, including nozzle temperature, extrusion flow rate, printing speed, cooling fan speed, platform temperature, and interlayer cooling time, and calculate their first and second derivatives to obtain the parameter change trends. For example, calculate the time derivative of the nozzle temperature to determine whether the temperature change rate is abnormal.Secondly, a time series anomaly detection method is adopted, such as a time series comparison algorithm based on dynamic time warping (DTW), to analyze the similarity between the current printing data and the normal printing data, and detect whether there is an abnormal pattern. For example, if the deviation of the extrusion flow rate in a certain printing exceeds 15% compared with the normal printing data, it can be determined that there is an abnormal fused deposition state during this period. In addition, statistical methods are used to analyze the correlation between various parameters, such as calculating the Pearson correlation coefficient between the extrusion flow rate and the nozzle temperature, to determine whether there is an abnormal mismatch. For example, under normal circumstances, the correlation coefficient between the nozzle temperature and the extrusion flow rate should be above 0.8. If it is lower than 0.6, there is an abnormal fused deposition. Finally, a threshold-based method is used to classify the abnormal states. For example, temperature anomalies, extrusion anomalies, and interlayer cooling anomalies are classified separately and stored in the abnormal state data table for subsequent abnormal analysis and printing failure detection.

[0073] Step S2 includes the following steps:

[0074] Step S21: Obtain the state data of the thermoplastic filamentous material;

[0075] Step S22: Based on the fused deposition modeling abnormal state data, perform printing abnormal form deformation recognition to obtain printing abnormal form deformation data;

[0076] Step S23: According to the state data of the thermoplastic filamentous material, perform reverse inference on the abnormal wire drawing mechanism of the printing abnormal form deformation data to obtain the abnormal wire drawing mechanism;

[0077] Step S24: Based on the printing abnormal form deformation data and the abnormal wire drawing mechanism, estimate the probability of printing interlayer warping to obtain printing interlayer warping probability data;

[0078] Step S25: Perform cracking risk calculation on the printing interlayer warping probability data to obtain printing interlayer cracking risk data.

[0079] As an example of the present invention, refer to Figure 2 As shown, in this example, step S2 includes:

[0080] Step S21: Obtain the state data of the thermoplastic filamentous material;

[0081] In the embodiments of the present invention, the acquisition of the state data of the thermoplastic filamentous material involves real-time monitoring of the physical and thermodynamic properties of the material, including key parameters such as the melting temperature, viscosity, tensile strength, cooling shrinkage rate, and extrusion rate of the material. During the data acquisition process, the nozzle temperature is measured by a thermocouple sensor, and the pulse number of the extruder stepping motor is recorded in combination with an optical encoder to calculate the actual material extrusion rate. In addition, a high-precision strain gauge is used to monitor the tension exerted on the material during the stretching process, and an infrared thermal imager is used to measure the temperature distribution of the material after it is extruded from the nozzle. In the experiment, PLA (polylactic acid) is selected as the test material, and the viscosity change curve of the material is recorded at different temperatures (190°C, 200°C, 210°C, 220°C) and extrusion rates (5 mm / s, 10 mm / s, 15 mm / s), and the relationship between the tensile strength and the temperature is calculated. The acquired data is stored in the format of a time series database for subsequent analysis.

[0082] Step S22: Based on the abnormal state data of fused deposition modeling, identify the abnormal form deformation during printing to obtain the abnormal form deformation data;

[0083] In the embodiments of the present invention, during the process of identifying the abnormal form deformation during printing based on the abnormal state data of fused deposition modeling, first analyze the interlayer deposition trajectory, and extract key parameters such as the contour deviation, deposition width, and layer height uniformity of each layer. The printing process is photographed by a high-speed camera, and the abnormal form on the printing path is identified by using an edge detection algorithm. The Canny edge detection method is used to extract the contour of the molten filament and calculate its deviation from the theoretical path. For the area with a large deviation, the morphological analysis method is further used to calculate the area, length, and curvature distribution of the abnormal area. For example, when a deviation of more than 0.3 mm is detected in the deposition trajectory of a certain layer at a printing speed of 200 mm / s, it can be determined that there is an abnormal form deformation in this layer. After the abnormal data is stored, statistical analysis is used to determine whether it belongs to a systematic abnormality, such as insufficient extrusion, excessive retraction, and overcooling shrinkage.

[0084] Step S23: Based on the state data of the thermoplastic filamentous material, inversely deduce the abnormal wire drawing mechanism for the abnormal form deformation data during printing to obtain the abnormal wire drawing mechanism;

[0085] In the embodiments of the present invention, the reverse deduction of the abnormal wire drawing mechanism is based on the analysis of the thermoplastic characteristics and rheological properties of the material, and the mechanism of wire drawing is deduced by using the material state data and the abnormal shape deformation data of printing. First, calculate the thermal viscosity drop gradient of the hot-melt filamentous material, use the Arrhenius equation to describe the viscosity change trend of the material, and fit the temperature-viscosity curve of PLA in combination with experimental data. Secondly, simulate the deformation behavior of the material based on the change of the thermoplastic fracture elongation rate, and use the finite element analysis (FEA) method to calculate the tensile strength and fracture elongation rate of the material at different temperatures. For example, at 190 °C, the fracture elongation rate of PLA is 3.2%, while it increases to 5.8% at 220 °C. Then, analyze the contour parameters of abnormal deformation during printing, such as wire drawing length and diameter change, and calculate the thermoplastic elongation rate distribution of local abnormal deformation. Combining the thermoplastic characteristics of the material, calculate the abnormal printing retraction distance and speed based on the Newton-Raphson method, so as to deduce the formation mechanism of abnormal wire drawing.

[0086] Step S24: Estimate the probability of warping between printing layers based on the abnormal shape deformation data of printing and the abnormal wire drawing mechanism, and obtain the probability data of warping between printing layers;

[0087] In the embodiments of the present invention, the probability estimation of warping between printing layers is based on the abnormal shape deformation data and the abnormal wire drawing mechanism, and the probability assessment is carried out by calculating the internal cavity distribution, the change of the hot-melt tension of the material and the increment of the cooling shrinkage rate. First, use the X-ray CT scanning technology to analyze the distribution of internal cavities in the printed part, and calculate the influence of the cavity rate on the interlayer bonding strength. Secondly, based on the hot-melt tension model of the material, use the tensile test data to fit the relationship curve of the hot-melt tension changing with temperature, and combine the abnormal wire drawing mechanism to calculate the change of the hot-melt tension under different printing conditions. For example, at 220 °C, the hot-melt tension of PLA drops to 0.12 MPa, while it is 0.25 MPa at 200 °C. Finally, combine the cavity rate, the change of the hot-melt tension and the increment of the cooling shrinkage rate, calculate the probability of warping between layers through multiple linear regression analysis, and store the analysis results.

[0088] Step S25: Calculate the cracking risk of the probability data of warping between printing layers to obtain the cracking risk data of warping between printing layers.

[0089] In the embodiments of the present invention, the calculation of the risk of interlayer cracking in printing is based on the probability data of interlayer warping, comprehensively evaluating the risk of internal stress accumulation, cooling rate distribution, and instability of interlayer adhesion. First, the finite element method is used to calculate the interlayer internal stress distribution, establish a heat conduction and stress coupling model, and simulate the stress accumulation caused by temperature gradient during the cooling process of the material. Secondly, the non-uniform cooling rate of interlayer warping is evaluated, the interlayer temperature distribution is analyzed based on the thermal imager data, and the Fourier heat conduction equation is used to calculate the cooling rate difference between different layers. For example, at a platform temperature of 60°C, the cooling rate of the top layer reaches -15°C / s, while that of the middle layer is only -5°C / s, indicating a large temperature difference and resulting in interlayer stress instability. Finally, based on the internal stress numerical fluctuation data and the change of cooling rate, the risk interval analysis of adhesion instability is carried out, the Newton iteration method is used to calculate the interlayer stress mutation point, and the numerical calculation of the cracking risk is carried out. The calculated cracking risk data is used for the training of the subsequent risk identification model to improve the accuracy of print failure detection.

[0090] Step S23 includes the following steps:

[0091] Step S231: Analyze the thermal viscosity drop gradient of the hot-melt filament material state data to obtain the thermal viscosity drop gradient data;

[0092] Step S232: Based on the thermal viscosity drop gradient data, simulate the change of the thermoplastic fracture elongation rate of the hot-melt filament material state data to obtain the thermoplastic fracture elongation rate change data;

[0093] Step S233: Analyze the abnormal deformation contour parameters of the print abnormal morphology deformation data to obtain the abnormal deformation contour parameters;

[0094] Step S234: According to the thermoplastic fracture elongation rate change data, inversely calculate the local abnormal deformation thermoplastic elongation rate distribution of the abnormal deformation contour parameters to obtain the abnormal deformation thermoplastic distribution elongation rate;

[0095] Step S235: Based on the abnormal deformation thermoplastic distribution elongation rate and the abnormal deformation contour parameters, perform an abnormal calculation of the print retraction distance / speed to obtain the abnormal print retraction distance / speed;

[0096] Step S236: According to the abnormal print retraction distance / speed and the abnormal deformation thermoplastic distribution elongation rate, inversely deduce the abnormal wire drawing mechanism of the abnormal deformation contour parameters to obtain the abnormal wire drawing mechanism.

[0097] In the embodiments of the present invention, during the process of analyzing the thermal viscosity decline gradient of the state data of the hot-melt filamentous material, first, a thermal viscosity change curve is constructed based on the material viscosity data under different temperature conditions. The viscosity of the material at different temperatures (180°C, 190°C, 200°C, 210°C, 220°C) is measured using a Brookfield rotational viscometer, and the viscosity change values in the shear rate range from 10 s⁻¹ to 1000 s⁻¹ are recorded. The relationship between the thermal viscosity and temperature is fitted using the Arrhenius equation, and the thermal viscosity decline gradient of the material is calculated through differential operations. For example, in the PLA material, the viscosity measured at 200°C is 1850 mPa·s, and it drops to 1120 mPa·s at 210°C, and the calculated thermal viscosity decline gradient is approximately -73 mPa·s / °C. Based on the data changes in different temperature ranges, a thermal viscosity decline gradient data table is established and stored for subsequent calculations. During the process of simulating the change in the thermoplastic fracture elongation rate of the state data of the hot-melt filamentous material based on the thermal viscosity decline gradient data, first, the viscosity data within the corresponding temperature range are selected, and combined with the stress-strain experimental data of the material, the change in the fracture elongation rate at different temperatures is calculated. A uniaxial tensile experiment of the PLA material is carried out at 190°C, 200°C, 210°C, and 220°C using an Instron tensile testing machine, and the fracture elongation rate of the material at different strain rates is recorded. For example, at 200°C, the fracture elongation rate of PLA is measured to be 3.8%, and it increases to 5.2% at 220°C. Based on the experimental data, a thermoplastic fracture elongation rate change curve is constructed using the cubic spline interpolation method, and the change rate of the fracture elongation rate is calculated through numerical differentiation. The calculated thermoplastic fracture elongation rate change data are stored for subsequent abnormal deformation analysis. During the process of analyzing the abnormal deformation profile parameters of the printed abnormal shape deformation data, a high-speed camera and image processing technology are used to measure the abnormal shape that appears during the printing process. First, the Canny edge detection algorithm is used to identify the edges of the printing trajectory, and the deviation amplitude of the contour is calculated. Secondly, the profile parameters of the abnormal deformation are extracted through morphological analysis methods, including the maximum deviation, local curvature change, and deformation area. In the experiment, at a printing speed of 200 mm / s, it is found that the maximum deviation of the contour of a certain printing path reaches 0.35 mm, which is significantly larger than that of the normal path (within 0.1 mm). In addition, the local curvature distribution of the deformed area is calculated, and it is found that the curvature change in the abnormal area exceeds 0.05 mm⁻¹. All abnormal deformation profile parameters are stored as structured data for subsequent analysis. During the process of inversely calculating the local abnormal deformation thermoplastic elongation rate distribution of the abnormal deformation profile parameters based on the thermoplastic fracture elongation rate change data, first, the area affected by tension is extracted based on the spatial distribution of the abnormal deformation. Then, combined with the thermoplastic fracture elongation rate change data, the local tensile rate distribution of the abnormal area is calculated.The finite element method (FEM) is used to simulate the tensile deformation behavior of the abnormal area at different temperatures. The thermoplastic fracture elongation data of the material is input, and the corresponding boundary conditions are applied for iterative calculation. For example, at 220 °C, the local tensile rate of the abnormal area reaches 4.5%, which is significantly higher than that of the normal area (2.8%). Finally, the thermoplastic elongation rate distribution data of the abnormal deformation is obtained and stored for subsequent abnormal wire drawing analysis. During the process of calculating the abnormal printing retraction distance / speed based on the thermoplastic distribution elongation rate of the abnormal deformation and the abnormal deformation profile parameters, the critical condition for the material to retract after stretching is first calculated. According to the viscoelastic characteristics of the thermoplastic material, a retraction mechanical model is established, and the Maxwell viscoelastic equation is used to calculate the retraction distance and speed of the material at different temperatures and stretching rates. For example, at 200 °C, the measured retraction distance of PLA is 2.1 mm, while at 220 °C, the retraction distance increases to 3.6 mm. Combining the abnormal deformation profile parameters, the correlation between the retraction distance and the abnormal area is analyzed, and the retraction speed distribution is calculated. For example, in a specific abnormal area, the retraction speed reaches 0.8 mm / s, which is higher than 0.5 mm / s in the normal printing area. Finally, the abnormal printing retraction distance / speed data is generated and stored for reverse deduction of the abnormal wire drawing mechanism. During the process of reverse deducing the abnormal wire drawing mechanism for the abnormal deformation profile parameters based on the abnormal printing retraction distance / speed and the thermoplastic distribution elongation rate of the abnormal deformation, the relationship between the retraction rate and the stretching rate in the wire drawing area is first analyzed, and the boundary conditions for wire drawing formation are calculated. The piecewise linear regression method is used to fit the curve of the retraction distance and the stretching rate in the abnormal wire drawing area, and the wire drawing critical point is calculated. For example, during the printing of PLA material, when the retraction speed exceeds 0.75 mm / s and the local stretching rate exceeds 3.5%, the wire drawing phenomenon begins to appear. Combining the printing abnormal shape deformation data, the correlation between the wire drawing length and the nozzle movement trajectory is further analyzed, and the wire drawing length distribution and the change of the wire drawing diameter are calculated. In the experiment, at 200 °C, the measured wire drawing length is 4.2 mm, while at 220 °C, it increases to 6.8 mm. Based on all the calculated data, the abnormal wire drawing mechanism is deduced and stored for subsequent calculation of the warping probability and detection of printing failures.

[0098] Step S24 includes the following steps:

[0099] Step S241: Identify the internal cavity distribution of the printing abnormal shape deformation data to obtain the internal cavity distribution data of the abnormal shape;

[0100] Step S242: Reverse deduce the change of the material's hot melt tension based on the abnormal wire drawing mechanism to obtain the material's hot melt tension change data;

[0101] Step S243: Calculate the increment gradient of the hot melt cooling shrinkage rate according to the internal cavity distribution data of the abnormal shape and the material's hot melt tension change data to obtain the increment gradient of the hot melt cooling shrinkage rate;

[0102] Step S244: Perform multiple linear regression analysis on the increment gradient of the hot melt cooling shrinkage rate to obtain the linear regression gradient of the shrinkage rate increment;

[0103] Step S245: Estimate the warping probability between printing layers based on the linear regression gradient of the shrinkage rate increment to obtain the warping probability data between printing layers.

[0104] In the embodiments of the present invention, during the process of identifying the internal cavity distribution of the abnormal printing morphological deformation data, first, the internal structure data of the printed part is obtained by using industrial CT scanning technology, and image processing algorithms are used to identify and analyze the internal cavities. The X-ray tube voltage of the industrial CT scanning device is set to 100 kV, the exposure time is 500 ms, and the resolution of the obtained tomographic scan image is set to 50 μm. The density distribution of the printed part is stratified by using the Otsu threshold segmentation method, and the morphological filtering method is combined to remove noise. For the identified low-density regions, the connected region analysis algorithm is used to calculate the cavity area and distribution, and the cavity distribution data is stored. For example, in a certain PLA printed part, multiple internal cavities are identified, the maximum cavity area is 2.3 mm², the average cavity diameter is 0.8 mm, and they are mainly distributed at the intersections of the printing paths. Combining the cavity morphological data, the cavity volume and its distribution probability are further calculated for subsequent analysis. During the process of reverse deduction of the material hot melt tension change based on the abnormal wire drawing mechanism, first, the flow behavior of the material in the molten state during the printing process is analyzed, and the relationship between the nozzle extrusion rate and the tensile force is calculated. Using a high-temperature tensile test device, uniaxial tensile tests are carried out on PLA materials at three temperatures of 200 °C, 210 °C, and 220 °C, and the tension changes at different tensile rates are recorded. The experimental data shows that when the temperature is 200 °C and the tensile rate is 5 mm / s, the material tension is 2.1 N, and when the temperature is 220 °C and the tensile rate is 10 mm / s, the tension rises to 3.8 N. Based on these data, the non-linear fitting method is used to calculate the hot melt tension change rate, and the tension is reverse deduced in combination with the actual printing parameters in the abnormal wire drawing area. For example, in a certain abnormal wire drawing area, the extrusion speed is 50 mm / s and the nozzle temperature is 210 °C. The corresponding hot melt tension is calculated to be 3.2 N by interpolation and this data is stored for subsequent calculations. During the process of calculating the incremental gradient of the material hot melt cooling shrinkage rate based on the abnormal morphological internal cavity distribution data and the material hot melt tension change data, first, the shrinkage behavior of the material during the cooling process is calculated. The differential scanning calorimeter (DSC) is used to measure the thermal expansion coefficient of the material, the expansion rate is recorded in the range of 100 °C - 200 °C, and the finite element analysis software is used for thermal stress simulation to calculate the change of the cooling shrinkage rate. Taking PLA material as an example, during the process of cooling from 200 °C to room temperature, the measured shrinkage rate is 0.37%, but in the abnormal cavity area, due to the local reduction of the material density, the shrinkage rate increases to 0.45%. Combining the hot melt tension data, the incremental gradient of the shrinkage rate is calculated. For example, when the internal cavity density reaches 1.5 pieces / cm³, the local shrinkage rate increment reaches 0.08%. All calculation results are stored as data tables for subsequent analysis. During the process of performing multiple linear regression analysis on the incremental gradient of the hot melt cooling shrinkage rate, first, the shrinkage rate data under different printing conditions are collected, including parameters such as nozzle temperature, printing speed, layer thickness, internal cavity density, etc., and a regression analysis matrix is constructed.The relationship between the shrinkage rate increment and various parameters is regression-fitted using the least squares method. For example, under the conditions of 200 °C, a printing speed of 60 mm / s, a layer thickness of 0.2 mm, and an internal void density of 1.5 voids / cm³, the regression analysis yields a shrinkage rate increment prediction formula: ΔS = 0.005T + 0.002V + 0.01D - 0.0003H, where T is the nozzle temperature (°C), V is the printing speed (mm / s), D is the void density (voids / cm³), and H is the layer thickness (mm). Finally, the linear regression gradient data of the shrinkage rate increment is obtained and stored for calculating the warping probability. During the process of estimating the warping probability between printing layers based on the linear regression gradient of the shrinkage rate increment, first, the local stress distribution between different layers of the printed part is calculated according to the regression gradient, and the thermal stress change between the printing layers is simulated using the linear elastic finite element method. An interlayer adhesion model is established using ABAQUS software, with the Young's modulus of the material set to 3.5 GPa and the Poisson's ratio to 0.36. The temperature change condition is applied, and the interlayer warping risk is calculated. For example, in the corner area of a certain printed part, since the local shrinkage rate increment reaches 0.09%, the calculated interlayer peeling stress is 2.1 MPa, exceeding the interlayer bonding strength (1.8 MPa) of the PLA material. Therefore, the warping probability is as high as 85%. Finally, the warping probability data between the printing layers is calculated and stored for subsequent printing failure detection.

[0105] Step S25 includes the following steps:

[0106] Step S251: Evaluate the non-uniform cooling rate of the warping probability data between the printing layers to obtain the non-uniform cooling rate of interlayer warping;

[0107] Step S252: Simulate the numerical fluctuation of internal stress accumulation of the warping probability data between the printing layers according to the non-uniform cooling rate of interlayer warping to obtain the numerical fluctuation data of interlayer internal stress;

[0108] Step S253: Analyze the instability risk interval of the interlayer adhesion force based on the numerical fluctuation data of the interlayer internal stress and the non-uniform cooling rate of interlayer warping to obtain the instability risk interval of the interlayer adhesion force;

[0109] Step S254: Calculate the cracking risk based on the instability risk interval of the interlayer adhesion force to obtain the cracking risk data between the printing layers.

[0110] In the embodiments of the present invention, during the process of evaluating the non-uniform cooling rate of the warping probability data between printing layers, first, the temperature change of the printing layer is monitored in real time based on an infrared thermal imager, and the cooling rates at different positions are calculated. Using a FLIRA655sc infrared thermal imager, the temperature field data of the printing layer is recorded at a sampling frequency of 10 Hz during the printing process, and the temperature gradient change of the printing layer at different time points is analyzed. The second-order finite difference method is used to calculate the cooling rate, and a spatial coordinate system is established to map the cooling rate to different regions between the printing layers. For example, in the corner region of a certain printed part, the cooling rate is 0.12 °C / s, while in the middle region, the cooling rate is only 0.05 °C / s. Through the evaluation of the non-uniform cooling rate, the temperature gradient distribution between the layers is calculated, and the data is stored for subsequent analysis. During the process of numerically simulating the internal stress accumulation fluctuation of the warping probability data between the printing layers according to the non-uniform cooling rate of the interlayer warping, first, the change in internal stress caused by thermal shrinkage during the interlayer cooling process is calculated using the thermal stress-structure coupling simulation method. Using the ABAQUS finite element analysis software, a temperature field-stress field coupling model between the printing layers is constructed, and the thermal expansion coefficient of the PLA material is set to , the Young's modulus is 3.5 GPa and the Poisson's ratio is 0.36. During the cooling stage, temperature change boundary conditions are applied to calculate the stress accumulation rate in different regions. For example, in the edge region of the printed part, due to the high cooling rate, the peak value of the interlayer internal stress calculated reaches 4.2 MPa, while in the central region, it is only 2.3 MPa. Further, the time-domain Fourier transform (FFT) method is used to analyze the stress numerical fluctuation characteristics, extract the frequency components of the interlayer stress change, and store the results for subsequent risk analysis. During the process of analyzing the risk interval of interlayer adhesion instability based on the interlayer internal stress numerical fluctuation data and the non-uniform cooling rate of interlayer warping, first, the extreme value distribution of the interlayer shear stress is calculated, and the safety interval of the interlayer adhesion is analyzed based on the critical instability theory. The interlayer bonding strength is calculated using the friction slip model, and the threshold value of the interlayer shear strength of the PLA material is set to 2.1 MPa. The extreme value statistical method is used to calculate the peak value distribution of the interlayer shear stress under different cooling rates. For example, in a certain printing area, the maximum shear stress is 2.3 MPa, which exceeds the bonding threshold of the material, so it is marked as a high-risk area. At the same time, the Monte Carlo random simulation method is used to calculate the interlayer adhesion instability probability under different printing conditions. For example, when the cooling rate changes in the range of 0.08 °C / s to 0.15 °C / s, the instability probability reaches 72%. Finally, the risk interval data of interlayer adhesion instability are output and stored for crack risk calculation. During the process of calculating the crack risk based on the risk interval of interlayer adhesion instability, first, the critical stress intensity factor of interlayer cracking is calculated based on the fracture mechanics theory, and the crack propagation risk is evaluated in combination with the interlayer internal stress fluctuation characteristics. The Griffith fracture criterion is used to calculate the critical stress intensity factor of the PLA material, and the initial crack length is set to 100 μm, and the critical stress intensity threshold is calculated to be 0.72 MPa·m¹ / ². Further, the extended finite element method (XFEM) is used to simulate the interlayer crack propagation process and calculate the crack growth rate. For example, in the high warping area, the simulated crack propagation rate is 0.005 mm / s, while in the low warping area, the crack propagation rate is only 0.001 mm / s. Combining the crack propagation rate and the interlayer stress fluctuation characteristics, the overall cracking risk of the printed layer is calculated, and finally, the cracking risk data of the printed layer are generated and stored for use in the subsequent printing failure detection system.

[0111] Step S253 includes the following steps:

[0112] Numerically quantify the growth trend of the interlayer internal stress numerical fluctuation data to obtain the growth trend numerically quantified data;

[0113] Identify the distribution of monotonically increasing mutation points of the interlayer internal stress numerical fluctuation data according to the growth trend numerically quantified data to obtain the internal stress increasing mutation point distribution data;

[0114] Couple the non - linear relationship according to the distribution data of the internal stress increasing mutation points and the non - uniform cooling rate of the interlayer warping to obtain the coupling data of the internal stress mutation - cooling rate relationship;

[0115] Perform piece - wise local trend convergence on the coupling data of the internal stress mutation - cooling rate relationship to obtain the local convergence trend of the internal stress mutation - cooling rate;

[0116] Based on the Newton iteration method, perform local convergence numerical iteration on the local convergence trend of the internal stress mutation - cooling rate to obtain the convergence iteration numerical value of the internal stress mutation - cooling rate;

[0117] Conduct an analysis of the instability risk interval of the interlayer adhesion force according to the convergence iteration numerical value of the internal stress mutation - cooling rate to obtain the instability risk interval of the interlayer adhesion force.

[0118] In the embodiments of the present invention, in the process of numerically quantifying the growth trend of the interlayer internal stress numerical fluctuation data, first, based on the time-series stress data, calculate its growth rate within different time windows, and use the numerical differentiation method to obtain the numerical expression of the growth trend. The central difference method is used to calculate the derivative of the stress fluctuation data at each time step. Set the time step to 0.5 s, and calculate the internal stress growth rate in each time interval. For example, the stress data in a certain area is [2.1 MPa, 2.3 MPa, 2.7 MPa, 3.2 MPa], and the growth trend data [0.2 MPa / s, 0.4 MPa / s, 0.5 MPa / s] is obtained through numerical differentiation. Further, the moving average filter is used to smooth the growth trend data to eliminate local abnormal disturbances, and finally, stable growth trend numerical quantification data is obtained. In the process of identifying the distribution of monotonically increasing mutation points of the interlayer internal stress numerical fluctuation data according to the growth trend numerical quantification data, first calculate the second derivative of the stress growth curve to determine the position of the mutation point. Lagrange interpolation is used to fit the growth trend data, and the curvature change rate at each time point is calculated. When the second derivative exceeds the set threshold (for example, 0.1 MPa / s²), it is marked as a mutation point. For example, in the stress growth curve of a certain area, mutation points are detected at 2 s and 4.5 s, indicating that significant stress accumulation changes have occurred in this area at these time points. Finally, store the positions of all mutation points and the corresponding stress values to form the internal stress increasing mutation point distribution data. In the process of coupling the non-linear relationship between the internal stress increasing mutation point distribution data and the non-uniform cooling rate of the interlayer warping, the Pearson correlation coefficient is used to calculate the linear correlation between the two, and support vector regression (SVR) is used for non-linear regression fitting. Set the kernel function to the radial basis function (RBF), and the regularization parameter C = 1.0 to fit the relationship curve between the internal stress mutation points and the cooling rate. For example, in the corner area of a certain printed part, the stress mutation point distribution data is [3.2 MPa, 3.5 MPa, 3.9 MPa], and the corresponding cooling rate data is [0.10 °C / s, 0.12 °C / s, 0.15 °C / s]. A non-linear coupling equation is obtained through SVR fitting, and finally, the internal stress mutation-cooling rate relationship coupling data is formed. In the process of segmental local trend convergence of the internal stress mutation-cooling rate relationship coupling data, first use the two-sample t-test to detect whether there is a significant difference in the stress growth trends in different cooling rate intervals. Divide the entire data set into several sub-intervals according to the cooling rate, such as [0.08 °C / s, 0.10 °C / s], [0.10 °C / s, 0.12 °C / s], etc. Calculate the mean value of the stress growth rate in each interval and conduct a hypothesis test. If the P value is less than 0.05, it is determined that there is a significant change in the trend within this interval. Finally, conduct trend convergence analysis on all intervals and store the local convergence trend data.In the process of numerically iterating the local convergence trend of the internal stress mutation-cooling rate based on the Newton iteration method, first, a non-linear equation of the cooling rate and stress growth is constructed, and the Newton iteration method is used to solve the root of the equation. The initial value is set as the average value of the cooling rate (e.g., 0.11 °C / s), the updated value of each step is calculated, and the error limit is used. As the convergence judgment condition. For example, in an iterative calculation, the initial value is set to 0.11, the next value calculated is 0.108, and the iteration continues to 0.107. After the error meets the convergence condition, the convergent iterative numerical value of the internal stress mutation-cooling rate is finally obtained. In the process of analyzing the instability risk interval of the interlayer adhesion force based on the convergent iterative numerical value of the internal stress mutation-cooling rate, first, the threshold value of the interlayer adhesion strength is calculated, and the critical point of instability occurrence is predicted based on the convergent iterative numerical value. The interlayer shear strength of the PLA material is set to 2.1 MPa, and the Bayesian probability analysis method is used to calculate the instability risk probability at different cooling rates. For example, when the cooling rate is 0.12 °C / s, the calculated interlayer instability probability is 65%. When the cooling rate is reduced to 0.09 °C / s, the instability probability drops to 28%. Finally, the data of the instability risk interval of the interlayer adhesion force is stored and used for subsequent cracking risk analysis. Among them, it should be explained that the "convergent iterative numerical value of the internal stress mutation-cooling rate" refers to the coupling relationship between the interlayer internal stress mutation and the cooling rate quantified by numerical calculation methods during the 3D printing process, and the iterative method is used to solve its final stable value. During the cooling process of the printed layer, the cooling rates in different regions are different, resulting in uneven accumulation of the interlayer internal stress and mutations in some regions. There is a complex non-linear relationship between such mutation points and the cooling rate. To accurately analyze this relationship, first, the growth trend of the interlayer internal stress fluctuation is identified by numerical methods, and the position and intensity of the mutation points are detected. Then, a mathematical model of the stress mutation point and the cooling rate is established using the non-linear regression method, and local trend analysis is carried out to judge its convergence characteristics. On this basis, the Newton iteration method is used to numerically optimize the local convergence trend of the internal stress mutation-cooling rate, that is, starting from the initial estimated value, through continuous iterative updates, it gradually approaches the final stable coupling value to ensure that the calculation results can reflect the actual influence of the cooling rate on the internal stress mutation. The finally obtained "convergent iterative numerical value" is a stable mathematical solution, representing the numerical characteristics of the interlayer internal stress mutation under different cooling rate conditions, and can be used to further predict the instability risk of the interlayer adhesion force.

[0119] In another embodiment, when numerically quantifying the growth trend of the interlayer internal stress numerical fluctuation data, it is first necessary to perform a time series analysis on the data to identify the trend of internal stress fluctuations existing in the data. By performing a moving average process on the time series data, noise interference is removed, and the basic trend of the fluctuations is retained. The slope calculation method is used to quantify the growth change amount between data points to determine the growth rate at each time point. Through this method, the obvious growth trend during the internal stress fluctuation process can be clearly extracted. A threshold is set, and when the growth amount exceeds this threshold, it is marked as important growth trend data. The data quantified by this growth trend will serve as the basis for subsequent analysis, providing the growth rate of each data point and its variation law in the time series, and providing data support for subsequent mutation point detection and non-linear coupling. According to the growth trend numerically quantified data, when identifying the distribution of monotonically increasing mutation points in the interlayer internal stress numerical fluctuation data, first analyze the quantified growth trend data to find the monotonically increasing part in the data. Through differential calculation or a mutation point detection algorithm based on local extrema, identify the positions where the growth trend in the data undergoes mutations. The basis for identifying mutation points is that within a certain interval, when the growth trend shows a rapid rise or fall and the change rate exceeds the set threshold, it is marked as a mutation point. Through the sliding window algorithm, ensure that the internal stress mutations changing with time can be dynamically tracked. During the identification process of the mutation point distribution, consider the long-term and short-term dependencies of the time series to avoid misidentification or missed identification. The finally obtained data of the internal stress increasing mutation point distribution provides the positions and intensities of the mutations during the internal stress fluctuation process, further providing key data support for the non-linear coupling analysis of the internal stress and the cooling rate. When coupling the non-linear relationship based on the data of the internal stress increasing mutation point distribution and the non-uniform cooling rate of the interlayer warpage, first, it is necessary to correlate the position of the internal stress mutation point with the change in the cooling rate. Through the interpolation method, obtain the cooling rate data corresponding to different time points, especially the change rate of the cooling rate. Then, use the polynomial regression method to establish the non-linear relationship between the internal stress mutation and the cooling rate, and select an appropriate order for fitting to capture the complex mutual influence between the two. To enhance the accuracy of the model, introduce the regularization method to avoid overfitting. During this process, the occurrence of internal stress mutations and the change in the cooling rate have a significant mutual influence. The coupling analysis can effectively reveal the sensitivity of internal stress mutations under different cooling conditions and provide a detailed data model for subsequent risk analysis. When performing piecewise local trend convergence on the coupled data of the internal stress mutation - cooling rate relationship, first divide the coupled data into several sub-intervals according to the change interval of the cooling rate, and the data within each sub-interval has similar cooling rate characteristics. Then, within each sub-interval, calculate the local convergence trend of the data, that is, fit the data change trend in each sub-interval through the local regression analysis method and evaluate its change smoothness.Local trend convergence analysis helps to determine whether there is a certain regular trend convergence in the occurrence of internal stress mutations under specific cooling rate conditions. During the analysis process, a window function is used to optimize the convergence process and ensure a high degree of trend convergence in each sub-interval, thus providing stable initial conditions for subsequent numerical iterations. When performing local convergence numerical iteration on the local convergence trend of internal stress mutation - cooling rate based on the Newton iteration method, the Newton iteration method is used to refine and optimize the local trend. Set the initial convergence value and calculate its deviation, and improve the accuracy by gradually approaching the true value. In each iteration, based on the numerical value of the current convergence trend, calculate the change rate of the local trend through derivation, and continuously adjust the convergence value until the set accuracy standard is met. An appropriate step size needs to be selected during the iteration process to ensure the stability of the calculation process and avoid numerical divergence or non-convergence of the calculation caused by too large a step size. Through multiple iterations, finally obtain stable internal stress mutation - cooling rate convergence iteration numerical values, providing accurate data support for subsequent risk interval analysis. When performing interlayer adhesion instability risk interval analysis based on the internal stress mutation - cooling rate convergence iteration numerical values, first construct a risk analysis model based on the converged numerical values. Using statistical analysis methods, model the change in adhesion force corresponding to the convergence values of internal stress mutation and cooling rate. Analyze the stability of the adhesion force under different internal stress levels through the non-linear relationship between interlayer adhesion force and internal stress mutation - cooling rate. When the coupling value of internal stress mutation and cooling rate exceeds the set instability threshold, predict the risk interval of interlayer adhesion instability. At this time, through interval analysis methods, clarify the critical values of interlayer adhesion instability under different internal stress and cooling rate conditions. The finally obtained interlayer adhesion instability risk interval provides an early warning for failures occurring during the 3D printing process, helps to monitor and adjust parameters in real time during the printing process, thereby improving the printing quality and success rate.

[0120] Step S3 includes the following steps:

[0121] Step S31: Perform convolution calculation on the risk level of printing layer cracking to obtain printing layer cracking risk convolution data;

[0122] Step S32: Perform logical learning on the printing layer cracking risk convolution data to obtain printing layer cracking risk logical data;

[0123] Step S33: Based on a convolutional neural network, construct a printing layer cracking risk identification model for the printing layer cracking risk logical data to obtain a printing layer cracking risk identification model.

[0124] As an example of the present invention, refer to Figure 3 As shown, in this example, step S3 includes:

[0125] Step S31: Perform convolution calculation on the risk level of interlayer cracking during printing to obtain the convolution data of the risk of interlayer cracking during printing;

[0126] In the embodiment of the present invention, during the process of performing convolution calculation on the risk level of interlayer cracking during printing, first, obtain the risk data of interlayer cracking during printing, and discretize the data to construct a multi-dimensional tensor data set. Based on the material characteristics of 3D printing, set the thickness distribution, cooling rate, thermal expansion coefficient, and stress accumulation data of the interlayer during printing as input variables, and perform data normalization processing to make the data values distributed in the range of 0 to 1 to improve the calculation accuracy. Use a three-dimensional convolution kernel to perform convolution operations on the interlayer data, set the size of the convolution kernel to 3×3×3, and use a sliding window with a step size of 1 to scan the entire data set layer by layer. Extract the stress gradient change characteristics of different interlayers through convolution calculation, and use the ReLU activation function to truncate negative value data to ensure the effectiveness of maintaining non-linear characteristics during the calculation process. Finally, generate the convolution data of the risk of interlayer cracking during printing, which contains the risk distribution of different regions of the interlayer during printing and is used for subsequent logical analysis.

[0127] Step S32: Perform logical learning on the convolution data of the risk of interlayer cracking during printing to obtain the logical data of the risk of interlayer cracking during printing;

[0128] In the embodiment of the present invention, during the process of performing logical learning on the convolution data of the risk of interlayer cracking during printing, first convert the convolution data into a matrix representation, and use the decision tree method to perform classification learning on the data. Set the cracking risk categories of different interlayers, including low risk, medium risk, and high risk categories, and construct a decision tree model based on the existing cracking history data. Use the information gain algorithm to calculate the weights of different features, and select the feature with the highest information gain as the decision node. For example, in a certain batch of printing data, the correlation coefficient between the interlayer cooling rate and the cracking probability reaches 0.85, so this feature is preferentially selected into the decision path. Further use the C4.5 algorithm to prune the decision tree to remove redundant nodes and improve the generalization ability of logical learning. Finally, form the logical data of the risk of interlayer cracking during printing, which is used to construct a cracking risk identification model.

[0129] Step S33: Based on the convolutional neural network, construct a printing interlayer cracking risk identification model for the logical data of the printing interlayer cracking risk to obtain a printing interlayer cracking risk identification model.

[0130] In the embodiments of the present invention, in the process of constructing a printing interlayer cracking risk identification model based on a convolutional neural network for printing interlayer cracking risk logic data, first, a neural network structure including multiple convolutional layers and pooling layers is constructed, and the dimension of the input data is set to 100×100×10 to adapt to the distribution of different printing layers. The size of the convolutional kernel of the first convolutional layer is set to 5×5, and the number of channels is 16. The max pooling layer is used for feature dimensionality reduction to reduce the computational complexity. Subsequently, the second convolutional layer is introduced, with the convolutional kernel size set to 3×3 and the number of channels increased to 32, and the batch normalization method is used to accelerate the convergence process. In the fully connected layer stage, the Softmax activation function is used for risk probability calculation, and the model is optimized based on the cross-entropy loss function. The Adam optimizer is used to adjust the learning rate, with the initial learning rate set to 0.001 and decaying to 0.9 times the original value after every 10 rounds of training. Finally, a printing interlayer cracking risk identification model is obtained, which can predict the cracking risks of different interlayers based on the input printing parameters and interlayer stress data, and provide data support for quality control in the 3D printing process.

[0131] Step S33 includes the following steps:

[0132] Step S331: Extract multi-scale features from the printing interlayer cracking risk logic data to obtain interlayer cracking multi-scale risk data;

[0133] Step S332: Divide the interlayer cracking multi-scale risk data into a training set and a test set to obtain an interlayer cracking multi-scale risk training set and an interlayer cracking multi-scale risk test set respectively;

[0134] Step S333: Construct an initial printing interlayer cracking risk identification model based on the convolutional neural network for the interlayer cracking multi-scale risk training set to obtain an initial printing interlayer cracking risk identification model;

[0135] Step S334: Test the initial printing interlayer cracking risk identification model according to the interlayer cracking multi-scale risk test set to obtain a printing interlayer cracking risk identification model.

[0136] In the embodiments of the present invention, during the process of extracting multi-scale features of the logical data of the interlayer cracking risk in printing, first, feature analysis is performed on the data based on convolution kernels of different scales to extract the cracking risk features of different levels in the printing interlayer. The sizes of the convolution kernels are set to 3×3, 5×5, and 7×7 respectively, and the convolution calculation is performed on the data in a sliding window manner to extract local, regional, and global interlayer risk features. For 3D printing materials with different layer thicknesses, a dynamic pooling strategy is adopted to enable feature extraction to adapt to different printing conditions. For thinner interlayer regions, a smaller pooling window, such as a 2×2 pooling window, is used to retain more detailed information; for thicker interlayer regions, a 4×4 pooling window is used to reduce the data dimension and improve the calculation efficiency. In this way, multi-scale risk data of interlayer cracking is obtained. This data contains the risk change features of different scales during the printing process and provides data support for subsequent training and testing. During the process of dividing the multi-scale risk data of interlayer cracking into a training set and a testing set, first, the data set is randomized, and the method of stratified sampling is adopted to ensure the balanced distribution of data with different risk levels. During the division process, 80% of the data is set as the training set, 20% of the data is set as the testing set, and the statistical features of the training set and the testing set are kept consistent in terms of interlayer thickness, cooling rate, stress distribution, etc. The K-fold cross-validation method is adopted to further divide a validation set within the training set to ensure that the model has strong generalization ability during the training process. For abnormal data points, such as data with interlayer warping exceeding a set threshold (such as 0.5 mm) or local stress concentration exceeding the material yield limit (such as 50 MPa), data augmentation methods are adopted for expansion to enhance the model's recognition ability for extreme situations. Finally, a multi-scale risk training set of interlayer cracking and a multi-scale risk testing set of interlayer cracking are obtained, providing data input for subsequent model training and testing. During the process of constructing an initial printing interlayer cracking risk recognition model based on a convolutional neural network for the multi-scale risk training set of interlayer cracking, first, a deep convolutional neural network is constructed, including multiple convolutional layers, pooling layers, and fully connected layers. The input layer receives the normalized training data. The first convolutional layer uses 16 3×3 convolutional kernels, and the second convolutional layer uses 32 5×5 convolutional kernels to extract local and medium-scale risk features in the interlayer. Subsequently, a Batch Normalization layer is added to normalize the data of different batches to accelerate the training convergence speed. The pooling layer adopts the max pooling strategy to reduce the data dimension and improve the calculation efficiency. At the fully connected layer stage, the Dropout method is introduced, and the dropout rate is set to 0.5 to prevent the model from overfitting. Finally, the Softmax activation function is used to classify different risk levels, and the cross-entropy loss function is used to calculate the error. The Adam optimization algorithm is used to adjust the learning rate, and the initial learning rate is set to 0.001 and dynamically adjusted during the training process.After training, an initial model for identifying the cracking risk between printing layers is obtained. This model can preliminarily identify the cracking risk levels between different printing layers. During the process of testing the initial model for identifying the cracking risk between printing layers using the multi-scale risk test set for interlayer cracking, the test set data is first used as input, and the predicted results output by the model are compared and analyzed with the actually labeled risk levels. The classification accuracy of the model is calculated using a confusion matrix, and indicators such as Precision, Recall, and F1-score are used to evaluate the model performance. For misclassified samples, error analysis is performed, and the Grad-CAM method is used to visualize the feature extraction process of the convolutional neural network to observe the key points of attention of the model on different interlayer features. During the testing process, if it is found that the model has a high misclassification rate at a specific layer thickness or cooling rate, a data rebalancing method is adopted to increase the data volume of this category and perform incremental training. Through multiple rounds of testing and optimization, a model for identifying the cracking risk between printing layers is finally obtained. This model can accurately predict the cracking risk between printing layers and provide reliable data support for 3D printing failure detection.

[0137] The present invention also provides a 3D printer printing failure detection system for executing the 3D printer printing failure detection method as described above. This 3D printer printing failure detection system includes:

[0138] An abnormal state extraction module for obtaining the operation log of the 3D printer; extracting the abnormal state of fused deposition modeling from the operation log of the 3D printer to obtain the abnormal state data of fused deposition modeling;

[0139] A cracking risk calculation module for inversely deducing the abnormal wire drawing mechanism based on the abnormal state data of fused deposition modeling to obtain the abnormal wire drawing mechanism; estimating the probability of warping between printing layers based on the abnormal wire drawing mechanism to obtain the probability data of warping between printing layers; calculating the cracking risk between printing layers from the probability data of warping between printing layers to obtain the cracking risk data between printing layers;

[0140] A cracking risk identification model construction module for constructing a cracking risk identification model between printing layers based on the convolutional neural network for the cracking risk data between printing layers to obtain a cracking risk identification model between printing layers;

[0141] A detection execution module for sending the cracking risk identification model between printing layers to the 3D printer control center to execute printing failure detection.

[0142] The above are only specific embodiments of the present invention, enabling those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but rather to the broadest scope consistent with the principles and novel features invented herein.

Claims

1. A 3D printer printing failure detection method, characterized in that: The following steps are involved: Step S1: Obtaining the 3D printer operation log; Extract abnormal state of fused deposition modeling from the operation log of the 3D printer to obtain abnormal state data of fused deposition modeling; Wherein, step S2 comprises: Step S21: obtaining hot-melt filament material state data; Step S22: performing printing abnormal morphological deformation identification based on the fused deposition modeling abnormal state data to obtain printing abnormal morphological deformation data; Step S23: reversely inferring the abnormal wire drawing mechanism of the printed abnormal morphological deformation data according to the state data of the hot-melt filamentary material to obtain the abnormal wire drawing mechanism; Step S24: estimating the probability of warping between printing layers based on the printing abnormal morphological deformation data and the abnormal wire drawing mechanism, and obtaining the probability data of warping between printing layers; Step S25: performing cracking risk calculation on the probability data of warping between printed layers to obtain cracking risk data between printed layers; Step S3: constructing a printing layer crack risk identification model based on the printing layer crack risk data based on the convolutional neural network to obtain a printing layer crack risk identification model; Step S4: Sending the inter-layer crack risk identification model to the 3D printer control center to perform printing failure detection.

2. The 3D printer printing failure detection method according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: Obtaining the 3D printer operation log; Step S12: performing data cleaning on the 3D printer operation log to obtain a 3D printer operation cleaning log; Step S13: extracting abnormal state of fused deposition modeling from the cleaning log of the 3D printer to obtain abnormal state data of fused deposition modeling.

3. The 3D printer printing failure detection method according to claim 1, characterized in that: Step S23 includes the following steps: Step S231: performing a thermal viscosity drop gradient analysis on the state data of the hot-melt filamentary material to obtain thermal viscosity drop gradient data; Step S232: performing a thermoplastic elongation at break change simulation on the state data of the hot-melt filamentary material based on the thermal viscosity decrease gradient data to obtain thermoplastic elongation at break change data; Step S233: analyzing the abnormal deformation contour parameters of the printed abnormal morphological deformation data to obtain the abnormal deformation contour parameters; Step S234: performing local abnormal deformation thermoplastic elongation distribution inversion on the abnormal deformation contour parameters according to the thermoplastic elongation at break change data to obtain the abnormal deformation thermoplastic distribution elongation; Step S235: performing abnormal calculation of printing retraction distance / speed based on abnormal deformation thermoplastic distribution elongation and abnormal deformation profile parameters to obtain abnormal printing retraction distance / speed; Step S236: reversely infer the abnormal wire drawing mechanism of the abnormal deformation profile parameters according to the abnormal printing retraction distance / speed and the abnormal deformation thermoplastic distribution elongation to obtain the abnormal wire drawing mechanism.

4. The 3D printer printing failure detection method according to claim 1, characterized in that: Step S24 includes the following steps: Step S241: performing internal void distribution identification on the printed abnormal morphology deformation data to obtain abnormal morphology internal void distribution data; Step S242: reversely infer the change of hot melt tension of the material based on the abnormal wire drawing mechanism to obtain the change data of hot melt tension of the material; Step S243: Calculate the incremental gradient of the material hot melt cooling shrinkage rate according to the abnormal morphology internal void distribution data and the material hot melt tension change data to obtain the incremental gradient of the hot melt cooling shrinkage rate; Step S244: performing a multivariate linear regression analysis on the incremental gradient of the hot melt cooling shrinkage rate to obtain the incremental linear regression gradient of the shrinkage rate; Step S245: estimating the probability of warping between printing layers based on the linear regression gradient of the shrinkage increment to obtain the probability data of warping between printing layers.

5. The 3D printer printing failure detection method according to claim 1, characterized in that: Step S25 includes the following steps: Step S251: evaluating the non-uniform cooling rate of the inter-layer warping probability data to obtain the inter-layer warping non-uniform cooling rate; Step S252: performing numerical fluctuation simulation of internal stress accumulation on the probability data of interlayer warping according to the non-uniform cooling rate of interlayer warping to obtain numerical fluctuation data of interlayer internal stress; Step S253: performing interlayer adhesion instability risk interval analysis based on interlayer internal stress numerical fluctuation data and interlayer warping non-uniform cooling rate to obtain an interlayer adhesion instability risk interval; Step S254: performing cracking risk calculation according to the interlayer adhesion instability risk interval to obtain interlayer cracking risk data.

6. The 3D printer printing failure detection method according to claim 1, characterized in that: Step S3 includes the following steps: Step S31: performing convolution calculation on the risk level of cracking between printed layers to obtain convolution data of the risk of cracking between printed layers; Step S32: performing logic learning on the risk convolution data of cracking between printed layers to obtain logic data of cracking risk between printed layers; Step S33: constructing a printing layer crack risk identification model based on the printing layer crack risk logic data based on the convolutional neural network to obtain the printing layer crack risk identification model.

7. The 3D printer printing failure detection method according to claim 6, characterized in that: Step S33 includes the following steps: Step S331: extracting multi-scale features from the printed interlayer crack risk logic data to obtain interlayer crack multi-scale risk data; Step S332: dividing the interlayer cracking multi-scale risk data into a training set and a test set, and obtaining an interlayer cracking multi-scale risk training set and an interlayer cracking multi-scale risk test set respectively; Step S333: constructing an initial printing interlayer crack risk identification model for the interlayer crack multi-scale risk training set based on a convolutional neural network to obtain an initial printing interlayer crack risk identification model; Step S334: performing a model test on the initial printing interlayer crack risk identification model according to the interlayer crack multi-scale risk test set to obtain a printing interlayer crack risk identification model.

8. A 3D printer printing failure detection system, characterized in that: Used to execute the 3D printer printing failure detection method according to claim 1, the 3D printer printing failure detection system comprises: The abnormal state extraction module is used to obtain the operation log of the 3D printer; perform fused deposition modeling abnormal state extraction on the operation log of the 3D printer to obtain fused deposition modeling abnormal state data; The cracking risk calculation module is used to reverse the abnormal wire drawing mechanism based on the abnormal state data of fused deposition modeling to obtain the abnormal wire drawing mechanism; estimate the probability of warping between printing layers based on the abnormal wire drawing mechanism to obtain the probability data of warping between printing layers; perform cracking risk calculation on the probability data of warping between printing layers to obtain the risk data of cracking between printing layers; A cracking risk identification model building module is used to build a printing interlayer cracking risk identification model based on the printing interlayer cracking risk data based on a convolutional neural network to obtain a printing interlayer cracking risk identification model; The detection execution module is used to send the printing layer crack risk identification model to the 3D printer control center to perform printing failure detection.

Images